MANSY: Generalizing Neural Adaptive Immersive Video Streaming With Ensemble and Representation Learning

TL;DR

MANSY is a novel neural streaming system that uses ensemble and representation learning to improve viewport prediction and QoE across diverse user preferences in immersive video streaming.

Contribution

It introduces a Transformer-based viewport prediction model with implicit ensemble learning and combines representation learning with deep reinforcement learning for adaptive bitrate selection.

Findings

Outperforms state-of-the-art in viewport prediction accuracy

Achieves better QoE across trained and unseen user preferences

Demonstrates improved generalization in diverse viewing scenarios

Abstract

The popularity of immersive videos has prompted extensive research into neural adaptive tile-based streaming to optimize video transmission over networks with limited bandwidth. However, the diversity of users' viewing patterns and Quality of Experience (QoE) preferences has not been fully addressed yet by existing neural adaptive approaches for viewport prediction and bitrate selection. Their performance can significantly deteriorate when users' actual viewing patterns and QoE preferences differ considerably from those observed during the training phase, resulting in poor generalization. In this paper, we propose MANSY, a novel streaming system that embraces user diversity to improve generalization. Specifically, to accommodate users' diverse viewing patterns, we design a Transformer-based viewport prediction model with an efficient multi-viewport trajectory input output architecture based on implicit ensemble learning. Besides, we for the first time combine the advanced representation learning and deep reinforcement learning to train the bitrate selection model to maximize diverse QoE objectives, enabling the model to generalize across users with diverse preferences. Extensive experiments demonstrate that MANSY outperforms state-of-the-art approaches in viewport prediction accuracy and QoE improvement on both trained and unseen viewing patterns and QoE preferences, achieving better generalization.